Chroma key (aka “bluescreen” or “greenscreen”) systems are used to create an image that combines a background still or video image (high-quality computer-generated graphics, for example) with video capture of a real object, such as a human being. Perhaps the most well known example of the use of chroma key systems is that of a television weatherman standing in front of a computer-generated weather map explaining the weather, while manipulating various graphical and video displays that appear behind him to illustrate the weather. This technology is also widely used in the television and movie industry, such as to combine live action in a studio setting (e.g. a close-up of an actor driving an automobile) with a pre-recorded background image (e.g. a street scene that could be visible through the rear window of the automobile).
A typical chroma key system 10 is illustrated in FIG. 1. A person 12 (or other real object) is positioned in front of a monochromatic surface 14, typically blue or green. The color of the surface is called the “chroma key.” A camera 16 captures an image of the person (or object), and this image provides one input into a chroma key processor 18.
A second input to the chroma key processor 18 is provided from a background image processor 20, which provides background image data, such as a high-quality computer-generated image. The chroma key processor combines the background image and the image of the real object by replacing the chroma key color with the background image. In the present example, the portion of the video image window that is occupied by the person 12 blocks the background image in that region, so that the background image does not appear. Wherever the chroma key color is detected, however, the background image persists. The result is a composite image 22 in which the person 12 appears to be standing in front of the high-quality computer-generated background image. This composite image is broadcast to viewers, who see only the final composite image, and may not even realize that it is a composite image.
In the real scene, however, the background image is not visible, and the background surface 14 appears to the person 12 (and other viewers of the real scene) as a featureless monochromatic surface. Hence, the person in the scene cannot directly interact with the background image. To provide feedback to the person, a monitor 24 is typically placed off-screen, and provided with the composite output 22 from the chroma key processor 18. By watching this monitor, the person can see their position in the composite image, such as to see where they appear to be pointing, etc. Unfortunately, interacting with the background image in a convincing way with this sort of system requires much practice, especially if the goal is to keep the viewer unaware of the whole process.